This is not possible, it is not supported, you need to use filebeat.yml
for those configurations, except the inputs.
For the inputs you can load its configuration from external config files as described in the documentation.
I have something like this.
The filebeat.yml
would look like this:
name: "node-name"
filebeat.config.inputs:
enabled: true
path: "/etc/filebeat/inputs/*.yml"
setup.ilm.enabled: false
ilm.enabled: false
setup.template.enabled: false
queue.mem:
events: 8000
flush.min_events: 500
flush.timeout: 1s
output.kafka:
output configuration
Then in the /etc/filebeat/inputs/
path I would put multiple yml
files with the inputs I want, something like this:
- type: filestream
paths:
- /path/to/file/*.json